{"id":28485783,"url":"https://github.com/abimathi03/face-prediction","last_synced_at":"2025-07-25T20:09:18.257Z","repository":{"id":289722801,"uuid":"867945262","full_name":"Abimathi03/Face-Prediction","owner":"Abimathi03","description":"Face Mask Reconstruction model using a Convolutional Autoencoder","archived":false,"fork":false,"pushed_at":"2025-04-24T17:46:09.000Z","size":682,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-29T10:39:34.563Z","etag":null,"topics":["autoencoder","convolution","deep-learning","facemask","model","reconstruction"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Abimathi03.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-10-05T04:27:03.000Z","updated_at":"2025-04-25T07:34:21.000Z","dependencies_parsed_at":"2025-04-24T18:47:44.335Z","dependency_job_id":"09eb2e13-c584-497c-bcc4-eea8e4f392cf","html_url":"https://github.com/Abimathi03/Face-Prediction","commit_stats":null,"previous_names":["abimathi03/face_prediction","abimathi03/face-prediction"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Abimathi03/Face-Prediction","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abimathi03%2FFace-Prediction","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abimathi03%2FFace-Prediction/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abimathi03%2FFace-Prediction/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abimathi03%2FFace-Prediction/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Abimathi03","download_url":"https://codeload.github.com/Abimathi03/Face-Prediction/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Abimathi03%2FFace-Prediction/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":267054629,"owners_count":24028338,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-07-25T02:00:09.625Z","response_time":70,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["autoencoder","convolution","deep-learning","facemask","model","reconstruction"],"created_at":"2025-06-08T00:40:43.063Z","updated_at":"2025-07-25T20:09:17.931Z","avatar_url":"https://github.com/Abimathi03.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🧠 Face Prediction using Masked Autoencoder\n\nThis project implements a **Face Mask Reconstruction** model using a **Convolutional Autoencoder**. It detects faces in an image, masks the lower half of the face, and reconstructs the masked area using a trained deep learning model. A custom loss function based on **Structural Similarity Index (SSIM)** is used to enhance the quality of reconstructions.\n\n---\n\n## 📑 Table of Contents\n\n1. [✨ Features](#-features)  \n2. [📦 Requirements](#-requirements)  \n3. [⚙️ Installation](#-installation)  \n4. [🚀 Usage of Intel OneAPI](#-usage-of-intel-oneapi)  \n5. [🧬 Code Overview](#-code-overview)  \n6. [📊 Visualization](#-visualization)  \n7. [📄 License](#-license)  \n8. [🙏 Acknowledgments](#-acknowledgments)\n\n---\n\n## ✨ Features\n\n- 📷 Upload an image and detect faces using **MTCNN**\n- 😷 Apply a mask to the lower half of the detected face\n- 🧠 Reconstruct the masked region using a **Convolutional Autoencoder**\n- 📊 Visualize the original, masked, and reconstructed images side-by-side\n\n---\n\n## 📦 Requirements\n\n- Python 3.x  \n- TensorFlow  \n- OpenCV  \n- Matplotlib  \n- NumPy  \n- dlib  \n- facenet-pytorch  \n\n---\n\n## ⚙️ Installation\n\n### 1. Clone the Repository\n\ngit clone https://github.com/Abimathi03/FACE_PREDICTION.git\ncd face-mask-reconstruction\n\n### 2. Install Dependencies\n\nInstall all required packages using pip:\n\npip install tensorflow opencv-python matplotlib numpy dlib facenet-pytorch\n\n⚠️ Ensure you have CMake and Visual Studio Build Tools (for Windows) for installing dlib.\n\n### 3. Download Pre-trained Models\n\nDownload the following file and place it in the project root:\n\n- shape_predictor_68_face_landmarks.dat (for facial landmark detection)\n\nYou can get it from: http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2\n\n## 🚀 Usage of Intel OneAPI\n\nIntel OneAPI is a toolkit for building optimized apps across CPUs and GPUs.\n\n### Steps:\n\n- Start a Jupyter Notebook or open a Google Colab session\n\n- Run the main script to upload and process an image:\n\n!python main.py\n\n- Follow the on-screen prompts to upload your image. The system will:\n\n- Detect the face\n\n- Apply a lower-face mask\n\n- Reconstruct the masked portion\n\n- Display results\n\n## 🧬 Code Overview\n### 🔍 Face Detection\n- Uses MTCNN to detect faces and facial landmarks.\n\n### 🧼 Image Preprocessing\n- Resize and normalize the input images.\n\n### 🏗️ Autoencoder Architecture\n- Encoder: Compresses input into a latent representation\n\n- Decoder: Reconstructs the image from the latent vector\n\n### 📐 Custom Loss Function\n- Uses SSIM (Structural Similarity Index) to compare masked vs reconstructed images for better visual fidelity.\n\n## 📊 Visualization\n\nThe system displays three outputs:\n\n- Original image with detected face\n\n- Masked face image (lower half masked)\n\n- Reconstructed image using autoencoder\n\nThis side-by-side view makes it easy to assess model performance.\n\n## 📄 License\n\nThis project is licensed under the MIT License.\nSee the LICENSE file for more information.\n\n## 🙏 Acknowledgments\n\n- MTCNN for face detection\n\n- TensorFlow for deep learning models\n\n- OpenCV for image processing utilities\n\n- dlib for facial landmarks detection\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fabimathi03%2Fface-prediction","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fabimathi03%2Fface-prediction","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fabimathi03%2Fface-prediction/lists"}